Management of Menopausal Symptoms in Breast Cancer Survivors

Daniela Stan

Dawn Hershman

Charles L. Loprinzi

INTRODUCTION

In the United States today, approximately 2.7 million women live as breast cancer survivors, a number that is increasing. This increase is particularly attributable to women diagnosed with early stage breast cancer, who have a life expectancy similar to age-matched controls. The issues facing cancer survivors are multiple and unique as the long-term side effects of cancer treatment and of aging play an increasingly prominent role in the routine care of these patients. Of special concern are the short and long-term effects of sex hormone deprivation such as vasomotor instability, osteoporosis, sexual dysfunction, arthralgias, and weight gain; currently there are several trials investigating ways to counteract these effects.

The average age of onset of natural menopause is 51 years. However, breast cancer therapy often leads to both an earlier onset of menopause and the exacerbation of existing menopausal symptoms. Hormone replacement therapy (HRT) is frequently used to treat these problems in the general population, but concerns about its potential to increase the risk of recurrence in breast cancer survivors have forced physicians to utilize alternative treatments.

While long-term HRT has beneficial effects on women’s bones and menopausal symptoms, this is more than offset by an increased risk of venous thromboembolic disease, breast cancer, stroke, cognitive decline, and coronary heart disease. In addition, the routine use of HRT in women does not appear to increase quality of life. These results support that long-term combination therapy with estrogen and progesterone cannot be recommended to most women at this time.

This chapter reviews current issues surrounding the acute and late effects associated with hormone deprivation in breast cancer survivors, and summarizes the scientific and therapeutic discoveries to date to identify optimal nonestrogenic treatments for individual patients.

VASOMOTOR INSTABILITY

Pathophysiology and Epidemiology

Hot flashes affect about three-fourths of all postmenopausal women, being the most commonly described health problem among this age group. Vasomotor instability usually begins 1 to 2 years prior to menopause and often persists for 6 months to 5 years after menopause. It is characterized by the sudden onset of a sensation of intense warmth that typically begins in the chest and progresses to the neck and face, often accompanied by anxiety, palpitations, profuse sweating, and red blotching of the skin. Hot flash symptoms can affect a woman’s ability to work, her social life, her sleep pattern, and her general perception of health and quality of life.

Changes in circulating estrogen levels can induce abnormalities of the central thermoregulatory centers, resulting in hot flashes. Perspiration and vasodilation, classic mechanisms of heat loss controlled by the hypothalamus, are activated during a hot flash. In normal homeostasis, these mechanisms are activated to maintain core body temperature in a regulated range termed the “thermoregulatory zone.” Complex neuroendocrine pathways that involve norepinephrine, estrogen, testosterone, and endorphins appear to govern regulation in the thermoregulatory nucleus and are possible sites where dysfunction may occur. In women undergoing natural menopause there is an association of hot flash symptoms with maternal history of hot flashes and with cigarette smoking.

Hot flashes are the most common reason women seek estrogen replacement therapy and, while estrogen effectively relieves symptoms for 80% to 90% of women who initiate treatment, women with a history of breast, ovarian, or uterine cancer, venous thromboembolism, coronary artery disease, or a family history of breast cancer comprise large populations for whom estrogen therapy may be contraindicated. Results from the Women’s Health Initiative (WHI) and the Million Women Study, in combination with other reports, suggest that long-term combined estrogen and progesterone therapy results in an increased risk of breast cancer, coronary heart disease, venous thromboembolism, and stroke. Although HRT with estrogen alone did not appear to increase the risk of breast cancer in the WHI study (1), the controversy surrounding its effects has been very influential in clinical practice and has led to a significant decrease in the use of HRT in the last decade (2). For these reasons, many women assume that hot flashes are an inevitable symptom of being a breast cancer survivor.

In breast cancer patients, vasomotor symptoms negatively impact sleep, quality of life, energy, as well as compliance with therapy and satisfaction with treatment decisions. The cause of vasomotor symptoms in breast cancer patients can be the result of abrupt estrogen loss due to surgery or chemotherapy, the use of adjuvant hormonal therapy, discontinuation of hormone replacement therapy, or from natural menopause. Tamoxifen, the most commonly prescribed pharmacologic treatment for breast cancer over the past decade, is associated with hot flashes in more than 50% of users (3, 4). Tamoxifen-associated hot flashes increase over the first several months of treatment and then gradually resolve (3). Postmenopausal women with a history of significant hot flashes prior to tamoxifen and a history of prior estrogen therapy use are likely to experience more severe hot flashes with tamoxifen therapy (3, 5). Compared to tamoxifen, adjuvant therapy with aromatase inhibitors (AIs) results in a lower incidence of hot flashes.

In premenopausal women with breast cancer, adjuvant chemotherapy is frequently associated with temporary or permanent amenorrhea, due to toxicity to the ovary. The incidence of chemotherapy-induced ovarian failure depends on the regimen used, the cumulative drug doses, and the age of the patient. The rapid changes in hormone concentrations associated with chemotherapy can lead to more severe symptoms than those of natural menopause.

Hormonal Therapies

Studies investigating the use of HRT for menopausal symptoms in breast cancer survivors have led to varying results. Although a few small pilot studies found the use of HRT after breast cancer safe, more recently, two randomized placebocontrolled studies of HRT—The HABITS study (6) and the LIBERATE study (7)—were closed early due to evidence of an increased breast cancer recurrence in the HRT group. The HABITS study of estrogen alone or in combination with progesterone versus placebo included 434 participants and closed at 2 years due to safety concerns, after a 3.5 relative hazard ratio for recurrence was seen in the HRT versus placebo group. In the LIBERATE study, 3,148 participants were randomized to tibolone (a synthetic compound with weak estrogenic, progestogenic, and androgenic actions) versus placebo; similar findings of increased breast cancer recurrence in the tibolone group were seen after a median follow-up of 3 years. In contrast to these findings, the Stockholm trial of 378 participants given estrogen with or without progestin versus placebo did not show a difference in recurrence after 4 years of follow-up (8). In this study the use of progesterone was minimized, which could explain the different results.

Some reports suggest that low-dose transdermal estrogen might be safer to use in this population and is very effective for symptom relief in the general population. In breast cancer survivors this therapy was not studied.

Current guidelines recommend that providers avoid the use of HRT for treatment of hot flashes in this population (9). If such therapy is used, it should be done over the shortest time period and with the lowest effective dose.

Progesterone alone, via oral megestrol acetate (20 mg/day) or the long-acting intramuscular depomedroxyproges-terone acetate (DMPA), was also demonstrated to be highly effective in the management of hot flashes in breast cancer survivors, with a 75% to 80% reduction in their severity and frequency (10, 11). Although minimal side effects are described during the treatment period, some women experienced withdrawal bleeding 1 to 4 weeks after discontinuation of treatment with megestrol acetate; the effects of the progesterone on hot flash reductions appear to be long-lasting. In a randomized trial comparing a single intramuscular dose of medroxyprogesterone acetate to venlafaxine, medroxyprogesterone acetate was more effective and appeared to have fewer short-term toxicities (12). It should be noted, however, that there are no long-term prospective data to establish the safety of progesterone analogs agents for women with hormone-sensitive breast cancer.

Nonhormonal Therapies

Newer Antidepressants

It is difficult to evaluate the efficacy of pharmacologic therapy for hot flashes with anecdotal reports alone, because of placebo effects. Multiple placebo-controlled trials demonstrate a 20% to 35% reduction in hot flashes with 4 weeks of placebo treatment. The most studied and effective nonhormonal pharmacological agents in the treatment of hot flashes are the newer antidepressants (selective serotonin reuptake inhibitors [SSRIs] and selective norepinephrine reuptake inhibitors [SNRIs]), as well as the anticonvulsants gabapentin and pregabalin.

Venlafaxine is thought to inhibit serotonin reuptake at lower doses and to induce a more profound inhibition of norepinephrine reuptake at higher doses. A double-blind, placebo-controlled trial with 191 breast cancer survivors randomized subjects to placebo or to one of three target venlafaxine doses (37.5 mg, 75 mg, or 150 mg daily) (13). After 4 weeks of treatment, the placebo groups had a 27% reduction in symptoms, versus 40%, 61%, and 61% reductions in the three venlafaxine groups, respectively. The side effects observed with venlafaxine include dry mouth, decreased appetite, nausea, and constipation (the latter only at doses of 150 mg/day).

A comparison study between venlafaxine and DMPA injections (12) showed a decrease in hot flashes by 55% with venlafaxine and by 79% with DMPA. In a crossover study comparing target dose venlafaxine ER 75 mg daily with gabapentin at a target dose of 300 mg three times daily, both treatments were equally effective (reduction in hot flashes scores by 66%), but venlafaxine was preferred by the patients (14). From these studies, it appears that venlafaxine adequately balances the symptom relief with the side effects. When venlafaxine is started, it should be started at 37.5 mg/day. When the drug is to be stopped, it should be slowly weaned down.

Another SNRI, desvenlafaxine, was found to be more effective than placebo, in two RCTs of women without cancer, at a target dose of 100 mg daily, with a starting dose of 50 mg daily for 3 days (15, 16).

SSRIs are also effective for the treatment of hot flashes. A double-blind, randomized, placebo-controlled, crossover clinical trial demonstrated that fluoxetine reduces the incidence of hot flashes, although the reduction does not appear to be as great as that observed with venlafaxine (13, 17). No significant toxicity was observed. Two placebo-controlled, double-blinded randomized trials found that paroxetine decreased hot flashes significantly more than did a placebo (18, 19 and 20). However, both paroxetine and fluoxetine are potent inhibitors of P450 (CYP) 2D6, an enzyme involved in the metabolism of tamoxifen to its active metabolite, endoxifen, thus raising concerns about the coadministration of these drugs with tamoxifen, as they may lower the mean endoxifen level. Endoxifen seems to also be the active metabolite in hot flashes, from studies showing that decreased CYP 2D6 activity leads to less prominent hot flashes after starting tamoxifen (and also worse breast cancer prognosis) and that patients without hot flashes on tamoxifen have an increase in breast cancer recurrence compared to those with hot flashes. Early studies supported that venlafaxine did not appear to inhibit CYP 2D6 to any appreciable degree; however, this remains a topic of investigation. Further information regarding tamoxifen metabolism can be found in Chapter 55.

Three randomized, placebo-controlled trials evaluating sertraline (50-100 mg/day) for the management of hot flashes failed to reveal any substantial benefit from this drug. Citalopram, a potent and specific SSRI, however, was shown to be effective against hot flashes in an RCT of 254 patients randomized to placebo, 10 mg, 20 mg, or 30 mg of citalopram daily (21). Hot flashes were reduced by 20%, 46%, 43%, and 50% respectively. Side effects were minimal. The recommended dose of citalopram is 10 to 20 mg daily. Escitalopram, a stereo-isomer of citalopram, was studied in 205 postmenopausal women and found to be more effective than placebo (47% vs. 33%) in decreasing the frequency of hot flashes (22). The dose used in this study was 10 mg/day. Increasing the does to 20 mg/day after 4 weeks was recommended in nonresponders.

Two randomized, placebo-controlled clinical trials of some of these newer antidepressants (23, 24) have been interpreted to be negative studies (25). Nonetheless, neither of these trials had hot flash diary data from a baseline period prior to the initiation of the study agents, making it impossible to determine how much hot flash reduction occurred. An individual patient pooled analysis of all the randomized trials looking at newer antidepressants that had baseline data available (26) demonstrates that they significantly decrease hot flashes (p < .0001).

Anticonvulsants and Other Centrally Acting Agents

Four placebo-controlled trials and a pooled analysis have confirmed the ability of gabapentin, a gamma aminobutyric acid (GABA) analogue, to decrease hot flashes (27, 28, 29, 30 and 31). Although one study suggested that gabapentin was as effective as estrogen therapy (29), this does not appear to be the case, as all data are considered together. The recommended starting gabapentin dose is 300 mg/day, titrating up to 900 mg/day. Side effects include lightheadedness and sleepiness early on, which generally resolve with continued treatment. Pregabalin also appears to be helpful for hot flash management, per a RCT of 75 mg twice a day versus 150 mg twice daily versus placebo (32). It is recommended at a starting dose is 50 mg/day, with a target dose of 75 mg twice a day.

Clonidine decreases hot flashes to a moderate degree but is associated with side effects that limit its utility.

Methyldopa and belladonna alkaloids do not appear to be very useful for hot flashes due to modest efficacy and unpleasant side effects.

Complementary and Alternative Agents

Despite anecdotal reports, the benefits of herbal therapies in clinical trials have been disappointing to date. Herbal treatments, such as black cohosh (Cimicifuga racemosa), ginseng (Panax Ginseng), evening primrose oil (Oenothera biennis), wild yam (Dioscorea villosa), and a standardized blend of 12 Chinese herbs have been prospectively evaluated, with minimal activity observed.

A randomized, placebo-controlled, crossover trial in 120 women found that vitamin E therapy decreased the average hot flash frequency by one episode per day (33), whereas more significant improvements in the vitamin E group were seen in a crossover trial (34). The low cost and minimal side effects of vitamin E make a trial of this agent (800 IU qd) one approach for individuals with mild symptoms that do not interfere with sleep or daily function. Therapy with vitamin E may increase the risk of heart failure in patients with vascular disease and diabetes mellitus but does not appear to affect the risk of cancers.

Studies of phytoestrogens in the treatment of hot flashes are also inconclusive. The isoflavone group (soy, red clover) was found to be no better than placebo in a meta-analysis of 17 placebo-controlled trials (25). Similarly, the lignants (found in flaxseed, whole grains, miller, fruits, and vegetables) did not help in hot flash treatment more than placebo in a randomized trial of flaxseed versus placebo bars (35). Furthermore, the long-term safety of pharmacologic doses of soy (90-400 mg/day) in patients with a history of breast or uterine cancer is not established.

The effect of nonestrogenic agents on hot flashes is similar in women with or without a history of breast cancer, irrespective of their use of tamoxifen (36).

Nonpharmacologic Interventions

Mind-body interventions, such as acupuncture, relaxationtraining/paced respiration, mindfulness-based stress reduction, cognitive behavioral therapy (CBT), and yoga all appear to be beneficial in small uncontrolled studies, but systematic reviews of some of these interventions have been inconclusive, given the small size of the studies and methodological flaws (37). Hypnosis has been suggested to be helpful via pilot trials and a small randomized trial of breast cancer survivors, published recently.

Although acupuncture appeared to be effective in alleviating hot flashes in uncontrolled studies, when compared to sham acupuncture, there was no significant reduction in symptoms. These findings were confirmed by a recent metaanalysis of 11 randomized trials (38), although some trials do still report positive results. There may be a form of acupuncture that does help hot flashes, but one has not been clearly identified, to date.

Another promising intervention in alleviating hot flashes in small studies is a stellate ganglion block; larger studies are underway to further investigate its effect. A review of RCTs on exercise for hot flashes was not able to demonstrate a benefit for this approach. For mild hot flashes, interventions such as using a fan, lowing room temperature, wearing loose-fitting clothing, consuming cold drinks, and avoiding alcohol and spices can be suggested.

Other comprehensive reviews of the pathogenesis and treatment of hot flashes in breast cancer survivors are available (39, 40 and 41). Table 51-1 provides a summary of the most active available therapies for hot flashes.

TABLE 51-1 Hot Flash Treatment Options

Treatment Option	Efficacy (%)^a	Comment
Estrogen	85%	Breast cancer concern
Progesterone^b	85%	Breast cancer concern
Antidepressants (e.g., venlafaxine, paroxetine, desvenlafaxine, citalopram, and escitalopram)	50%-60%	Do not use paroxetine or fluoxetine with tamoxifen
Gabapentin	50%-60%
^aPercentage of reduction from baseline after about 4 weeks, noting that a placebo decreases them about 20% to 30%. ^bMegestrol acetate 20 to 40 mg/d orally or medroxyprogesterone acetate 400 mg intramuscularly once.

OSTEOPOROSIS

Definition and Pathophysiology

Osteoporosis is a metabolic bone disease characterized by low bone mass and microarchitectural deterioration of bone tissue, leading to enhanced bone fragility and a consequent increase in fracture risk. Decreased body mass index (BMI) is an important known risk factor for low bone mineral density (BMD) and osteoporotic fractures. A BMD that is 2.5 standard deviations below the young adult female mean value is used to diagnose osteoporosis. At menopause, women enter a 10-year period of accelerated bone loss, responsible for a 20% to 30% loss of cancellous bone and a 5% to 10% loss of cortical bone. Estrogen and testosterone play important roles in the regulation of BMD and bone health. In postmenopausal women, the conversion of adrenal androgens to estrogen by the enzyme, aromatase, leads to continued low levels of circulating estrogen, which may play an important role in calcium homeostasis through its effects on renal and GI absorption of calcium. Osteoporotic fractures are an important health hazard; women have an estimated 1-year mortality risk of 21% after a hip fracture.

Characteristics in Breast Cancer Survivors

Women with premenopausal breast cancer have higher than average rates of bone loss and fracture, with nonmetastatic breast cancer patients having a risk of vertebral fractures nearly five times that of the general population. Premature ovarian failure due to chemotherapy, estrogen blocking therapies, or ovarian ablation therapy, as well as direct effects of chemotherapy on bone, inactivity, use of corticosteroids, and inadequate intake of calcium and vitamin D can all contribute to the increased osteoporosis risk in breast cancer survivors.

The effects of tamoxifen, a selective estrogen receptor modulator with both agonist and antagonist activity widely used as adjuvant endocrine therapy in hormone-receptor-positive breast cancer, on BMD differ depending on the menopausal status of the patient at the time of treatment. The first study to look at the bone effects of tamoxifen versus placebo was conducted by Love et al. in 1992, on 140 postmenopausal breast cancer survivors, over 2 years (42). The mean BMD at the lumbar spine increased by 0.61% per year in the tamoxifen group and decreased by 1% per year in the placebo group. Other studies in postmenopausal women have confirmed this observation (43, 44) whereas in premenopausal women tamoxifen therapy is associated with varying levels of bone loss. In a chemoprevention study of tamoxifen versus placebo in 179 women at high risk for breast cancer, premenopausal women treated with tamoxifen experienced a 1.4% decrease in BMD per year over the 3-year study, while the BMD increased 1.2% per year in postmenopausal women in the tamoxifen group (42, 45). In a population-based case-control study of women older than 50 who suffered a first osteoporotic fracture, the use of tamoxifen was associated with a lower risk of fractures compared to controls (46). Treatment with tamoxifen also reduces markers of bone turnover in postmenopausal women.

Only gold members can continue reading. Log In or Register to continue